Download quantization_manifest.json from tencent/WeMM-Video-4B-FP8: direct link, hf CLI and curl.
- Browser
- Download file 51.5 kB
-
https://huggingface.co/tencent/WeMM-Video-4B-FP8/resolve/main/quantization_manifest.json
- Command line
-
hf download hf://tencent/WeMM-Video-4B-FP8/quantization_manifest.json
-
curl -L -o quantization_manifest.json https://huggingface.co/tencent/WeMM-Video-4B-FP8/resolve/main/quantization_manifest.json
51.5 kB
{
"status": "quantized_and_independently_reloaded_native_inference_pending",
"algorithm": "Transformers 5.5.3 Fp8Quantize.convert",
"converter_sha256": "9baa1fdc55581355dabaf10d68ddafcd53983fe8736737c991c2ea78b03a3d9f",
"source": "tencent/WeMM-Video-4B",
"source_files": [
{
"path": "tencent/WeMM-Video-4B/model-00001-of-00003.safetensors",
"bytes": 4306667392,
"sha256": "43adb5f51324177c3c40661982f5729b2c927b8c2f3eba9352ece3e0d5b2e2fd"
},
{
"path": "tencent/WeMM-Video-4B/model-00002-of-00003.safetensors",
"bytes": 4331122608,
"sha256": "c271269cd9ad157ace651e2f6ee2f28c8ee80c52324f8979ddb138ba653f069c"
},
{
"path": "tencent/WeMM-Video-4B/model-00003-of-00003.safetensors",
"bytes": 2585872480,
"sha256": "b9ecafb62c8c758242c926830ed628ef685c42e0be6b779c05d1eff2da1fdfd9"
}
],
"source_config_sha256": "5eca5c5955cdc4ec20ddb433053130c690304a0a26c2dcd031daef032b7235a4",
"official_repo": "Qwen/Qwen3.5-35B-A3B-FP8",
"official_revision": "9d1823d2dee688a6b25e77009dc727688c44936e",
"recipe_sha256": "1fa272f54cc586827ffa0e6d0ce14805d298adab4c9bd78663967bef155efda7",
"files": {
"wemm-video-capabilities-compact.png": {
"bytes": 2718998,
"sha256": "57d892abb395c639078b49ecf317698f038303adf1fc976ddf13ff9e9273c77f"
},
"wemm-video-architecture.png": {
"bytes": 116917,
"sha256": "9902c081a70abeaacf34c898c86c44587eebe0fd62f1923225d1726766eb0a7a"
},
"chat_template.jinja": {
"bytes": 7755,
"sha256": "273d8e0e683b885071fb17e08d71e5f2a5ddfb5309756181681de4f5a1822d80"
},
"config.json": {
"bytes": 50625,
"sha256": "77201ba25b2b6655411d8a5df5c03b51d730d140f89f82c826e11e5105ee027e"
},
"configuration_wemm.py": {
"bytes": 37064,
"sha256": "43ee3ec8477c6677f06835a893a146d18024aea0145086ceca3b6039e2d91887"
},
"generation_config.json": {
"bytes": 138,
"sha256": "b91f22ecb30192b6919660eb61c01d7602f93ff6764556b1991c71fbe9ca82e6"
},
"h5_modeling_bert.py": {
"bytes": 50024,
"sha256": "b2eac797422294be10295c27f5709c8fda362c55d6f476dcebfd5f38aa65c6dd"
},
"h5_modeling_whisper.py": {
"bytes": 69582,
"sha256": "b507884daee13857c17d2d97290183e71a5a1044253a49c5a9cb109ba3332907"
},
"model.safetensors": {
"bytes": 7659403688,
"sha256": "bd5ed7eaf0682ee781175a249c9f86b9366d0d31980f72409ec6b7d939486ef4"
},
"modeling_wemm.py": {
"bytes": 103798,
"sha256": "30c9e34d16f4b3720410c47034b2b807a3686c88f4558d5aab6c102fc1db0bde"
},
"official_qwen35_fp8_config.json": {
"bytes": 19172,
"sha256": "cc35eadc02b2a69542c59233f5d80cd89029234711ac11811afc46d57701f7b9"
},
"preprocessor_config.json": {
"bytes": 390,
"sha256": "27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516"
},
"processing_wemm.py": {
"bytes": 24880,
"sha256": "8f907439e011c47293a851e28efec1501c36b862ba8c04784d23a0a4a5aa36a8"
},
"processor_config.json": {
"bytes": 2734,
"sha256": "7067f0990504250c037bc718650754575b9002133b247d510ca86bf475b30d29"
},
"tokenizer.json": {
"bytes": 19989343,
"sha256": "510fd6ff51823f6b1c820595b5eed750b8ff0ee85961d2f330623b3cf5ff4019"
},
"tokenizer_config.json": {
"bytes": 1098,
"sha256": "83febe739ecefd99f984f546dfdc10149cedf293b9ee0291b99894341188b613"
},
"video_preprocessor_config.json": {
"bytes": 385,
"sha256": "7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13"
},
"wemm_media.py": {
"bytes": 11378,
"sha256": "2e4b427e3ddabc526f7984066a982bfa7c0a4d56f3d7c979fc41651463461122"
},
"wemm_video_policy.py": {
"bytes": 5670,
"sha256": "5a08358f04b011c401d6355cf396caebae4536442df76ae17dc89470c2b1a50e"
},
"wemm_zip.py": {
"bytes": 12156,
"sha256": "016d7a2bc9b50cf2bed2e901d7ec445260eff890751e94683c8fac0be09eecdb"
},
"wemm_zip_policy.py": {
"bytes": 1625,
"sha256": "f4029f5015cf6342da1bfacaa6b4847d7f20ca450019ff7bc5f4ad899e3706c8"
},
"README.md": {
"bytes": 8083,
"sha256": "8ed9f6d18b2b9cf6d5c7ca81056fc39eb319555de63cf15d91bc923cac66ae1d"
},
"LICENSE": {
"bytes": 25380,
"sha256": "bac4d42ddebadb11e405fc4de2335579e67676362ac870108012ab4899ef1e5b"
}
},
"weights": "E4M3FN,128x128 blocks,FP32 weight_scale_inv",
"activations": "dynamic,per-token groups128",
"quantization_config": {
"quant_method": "fp8",
"activation_scheme": "dynamic",
"weight_per_tensor": false,
"act_per_tensor": false,
"weight_block_size": [
128,
128
],
"modules_to_not_convert": [
"lm_head",
"model.audio",
"model.audio.audio_proj",
"model.audio.audio_proj.bias",
"model.audio.conv1",
"model.audio.conv1.bias",
"model.audio.conv2",
"model.audio.conv2.bias",
"model.audio.embed_positions",
"model.audio.layer_norm",
"model.audio.layer_norm.bias",
"model.audio.layers.0.fc1",
"model.audio.layers.0.fc1.bias",
"model.audio.layers.0.fc2",
"model.audio.layers.0.fc2.bias",
"model.audio.layers.0.final_layer_norm",
"model.audio.layers.0.final_layer_norm.bias",
"model.audio.layers.0.self_attn.k_proj",
"model.audio.layers.0.self_attn.out_proj",
"model.audio.layers.0.self_attn.out_proj.bias",
"model.audio.layers.0.self_attn.q_proj",
"model.audio.layers.0.self_attn.q_proj.bias",
"model.audio.layers.0.self_attn.v_proj",
"model.audio.layers.0.self_attn.v_proj.bias",
"model.audio.layers.0.self_attn_layer_norm",
"model.audio.layers.0.self_attn_layer_norm.bias",
"model.audio.layers.1.fc1",
"model.audio.layers.1.fc1.bias",
"model.audio.layers.1.fc2",
"model.audio.layers.1.fc2.bias",
"model.audio.layers.1.final_layer_norm",
"model.audio.layers.1.final_layer_norm.bias",
"model.audio.layers.1.self_attn.k_proj",
"model.audio.layers.1.self_attn.out_proj",
"model.audio.layers.1.self_attn.out_proj.bias",
"model.audio.layers.1.self_attn.q_proj",
"model.audio.layers.1.self_attn.q_proj.bias",
"model.audio.layers.1.self_attn.v_proj",
"model.audio.layers.1.self_attn.v_proj.bias",
"model.audio.layers.1.self_attn_layer_norm",
"model.audio.layers.1.self_attn_layer_norm.bias",
"model.audio.layers.10.fc1",
"model.audio.layers.10.fc1.bias",
"model.audio.layers.10.fc2",
"model.audio.layers.10.fc2.bias",
"model.audio.layers.10.final_layer_norm",
"model.audio.layers.10.final_layer_norm.bias",
"model.audio.layers.10.self_attn.k_proj",
"model.audio.layers.10.self_attn.out_proj",
"model.audio.layers.10.self_attn.out_proj.bias",
"model.audio.layers.10.self_attn.q_proj",
"model.audio.layers.10.self_attn.q_proj.bias",
"model.audio.layers.10.self_attn.v_proj",
"model.audio.layers.10.self_attn.v_proj.bias",
"model.audio.layers.10.self_attn_layer_norm",
"model.audio.layers.10.self_attn_layer_norm.bias",
"model.audio.layers.11.fc1",
"model.audio.layers.11.fc1.bias",
"model.audio.layers.11.fc2",
"model.audio.layers.11.fc2.bias",
"model.audio.layers.11.final_layer_norm",
"model.audio.layers.11.final_layer_norm.bias",
"model.audio.layers.11.self_attn.k_proj",
"model.audio.layers.11.self_attn.out_proj",
"model.audio.layers.11.self_attn.out_proj.bias",
"model.audio.layers.11.self_attn.q_proj",
"model.audio.layers.11.self_attn.q_proj.bias",
"model.audio.layers.11.self_attn.v_proj",
"model.audio.layers.11.self_attn.v_proj.bias",
"model.audio.layers.11.self_attn_layer_norm",
"model.audio.layers.11.self_attn_layer_norm.bias",
"model.audio.layers.12.fc1",
"model.audio.layers.12.fc1.bias",
"model.audio.layers.12.fc2",
"model.audio.layers.12.fc2.bias",
"model.audio.layers.12.final_layer_norm",
"model.audio.layers.12.final_layer_norm.bias",
"model.audio.layers.12.self_attn.k_proj",
"model.audio.layers.12.self_attn.out_proj",
"model.audio.layers.12.self_attn.out_proj.bias",
"model.audio.layers.12.self_attn.q_proj",
"model.audio.layers.12.self_attn.q_proj.bias",
"model.audio.layers.12.self_attn.v_proj",
"model.audio.layers.12.self_attn.v_proj.bias",
"model.audio.layers.12.self_attn_layer_norm",
"model.audio.layers.12.self_attn_layer_norm.bias",
"model.audio.layers.13.fc1",
"model.audio.layers.13.fc1.bias",
"model.audio.layers.13.fc2",
"model.audio.layers.13.fc2.bias",
"model.audio.layers.13.final_layer_norm",
"model.audio.layers.13.final_layer_norm.bias",
"model.audio.layers.13.self_attn.k_proj",
"model.audio.layers.13.self_attn.out_proj",
"model.audio.layers.13.self_attn.out_proj.bias",
"model.audio.layers.13.self_attn.q_proj",
"model.audio.layers.13.self_attn.q_proj.bias",
"model.audio.layers.13.self_attn.v_proj",
"model.audio.layers.13.self_attn.v_proj.bias",
"model.audio.layers.13.self_attn_layer_norm",
"model.audio.layers.13.self_attn_layer_norm.bias",
"model.audio.layers.14.fc1",
"model.audio.layers.14.fc1.bias",
"model.audio.layers.14.fc2",
"model.audio.layers.14.fc2.bias",
"model.audio.layers.14.final_layer_norm",
"model.audio.layers.14.final_layer_norm.bias",
"model.audio.layers.14.self_attn.k_proj",
"model.audio.layers.14.self_attn.out_proj",
"model.audio.layers.14.self_attn.out_proj.bias",
"model.audio.layers.14.self_attn.q_proj",
"model.audio.layers.14.self_attn.q_proj.bias",
"model.audio.layers.14.self_attn.v_proj",
"model.audio.layers.14.self_attn.v_proj.bias",
"model.audio.layers.14.self_attn_layer_norm",
"model.audio.layers.14.self_attn_layer_norm.bias",
"model.audio.layers.15.fc1",
"model.audio.layers.15.fc1.bias",
"model.audio.layers.15.fc2",
"model.audio.layers.15.fc2.bias",
"model.audio.layers.15.final_layer_norm",
"model.audio.layers.15.final_layer_norm.bias",
"model.audio.layers.15.self_attn.k_proj",
"model.audio.layers.15.self_attn.out_proj",
"model.audio.layers.15.self_attn.out_proj.bias",
"model.audio.layers.15.self_attn.q_proj",
"model.audio.layers.15.self_attn.q_proj.bias",
"model.audio.layers.15.self_attn.v_proj",
"model.audio.layers.15.self_attn.v_proj.bias",
"model.audio.layers.15.self_attn_layer_norm",
"model.audio.layers.15.self_attn_layer_norm.bias",
"model.audio.layers.16.fc1",
"model.audio.layers.16.fc1.bias",
"model.audio.layers.16.fc2",
"model.audio.layers.16.fc2.bias",
"model.audio.layers.16.final_layer_norm",
"model.audio.layers.16.final_layer_norm.bias",
"model.audio.layers.16.self_attn.k_proj",
"model.audio.layers.16.self_attn.out_proj",
"model.audio.layers.16.self_attn.out_proj.bias",
"model.audio.layers.16.self_attn.q_proj",
"model.audio.layers.16.self_attn.q_proj.bias",
"model.audio.layers.16.self_attn.v_proj",
"model.audio.layers.16.self_attn.v_proj.bias",
"model.audio.layers.16.self_attn_layer_norm",
"model.audio.layers.16.self_attn_layer_norm.bias",
"model.audio.layers.17.fc1",
"model.audio.layers.17.fc1.bias",
"model.audio.layers.17.fc2",
"model.audio.layers.17.fc2.bias",
"model.audio.layers.17.final_layer_norm",
"model.audio.layers.17.final_layer_norm.bias",
"model.audio.layers.17.self_attn.k_proj",
"model.audio.layers.17.self_attn.out_proj",
"model.audio.layers.17.self_attn.out_proj.bias",
"model.audio.layers.17.self_attn.q_proj",
"model.audio.layers.17.self_attn.q_proj.bias",
"model.audio.layers.17.self_attn.v_proj",
"model.audio.layers.17.self_attn.v_proj.bias",
"model.audio.layers.17.self_attn_layer_norm",
"model.audio.layers.17.self_attn_layer_norm.bias",
"model.audio.layers.18.fc1",
"model.audio.layers.18.fc1.bias",
"model.audio.layers.18.fc2",
"model.audio.layers.18.fc2.bias",
"model.audio.layers.18.final_layer_norm",
"model.audio.layers.18.final_layer_norm.bias",
"model.audio.layers.18.self_attn.k_proj",
"model.audio.layers.18.self_attn.out_proj",
"model.audio.layers.18.self_attn.out_proj.bias",
"model.audio.layers.18.self_attn.q_proj",
"model.audio.layers.18.self_attn.q_proj.bias",
"model.audio.layers.18.self_attn.v_proj",
"model.audio.layers.18.self_attn.v_proj.bias",
"model.audio.layers.18.self_attn_layer_norm",
"model.audio.layers.18.self_attn_layer_norm.bias",
"model.audio.layers.19.fc1",
"model.audio.layers.19.fc1.bias",
"model.audio.layers.19.fc2",
"model.audio.layers.19.fc2.bias",
"model.audio.layers.19.final_layer_norm",
"model.audio.layers.19.final_layer_norm.bias",
"model.audio.layers.19.self_attn.k_proj",
"model.audio.layers.19.self_attn.out_proj",
"model.audio.layers.19.self_attn.out_proj.bias",
"model.audio.layers.19.self_attn.q_proj",
"model.audio.layers.19.self_attn.q_proj.bias",
"model.audio.layers.19.self_attn.v_proj",
"model.audio.layers.19.self_attn.v_proj.bias",
"model.audio.layers.19.self_attn_layer_norm",
"model.audio.layers.19.self_attn_layer_norm.bias",
"model.audio.layers.2.fc1",
"model.audio.layers.2.fc1.bias",
"model.audio.layers.2.fc2",
"model.audio.layers.2.fc2.bias",
"model.audio.layers.2.final_layer_norm",
"model.audio.layers.2.final_layer_norm.bias",
"model.audio.layers.2.self_attn.k_proj",
"model.audio.layers.2.self_attn.out_proj",
"model.audio.layers.2.self_attn.out_proj.bias",
"model.audio.layers.2.self_attn.q_proj",
"model.audio.layers.2.self_attn.q_proj.bias",
"model.audio.layers.2.self_attn.v_proj",
"model.audio.layers.2.self_attn.v_proj.bias",
"model.audio.layers.2.self_attn_layer_norm",
"model.audio.layers.2.self_attn_layer_norm.bias",
"model.audio.layers.20.fc1",
"model.audio.layers.20.fc1.bias",
"model.audio.layers.20.fc2",
"model.audio.layers.20.fc2.bias",
"model.audio.layers.20.final_layer_norm",
"model.audio.layers.20.final_layer_norm.bias",
"model.audio.layers.20.self_attn.k_proj",
"model.audio.layers.20.self_attn.out_proj",
"model.audio.layers.20.self_attn.out_proj.bias",
"model.audio.layers.20.self_attn.q_proj",
"model.audio.layers.20.self_attn.q_proj.bias",
"model.audio.layers.20.self_attn.v_proj",
"model.audio.layers.20.self_attn.v_proj.bias",
"model.audio.layers.20.self_attn_layer_norm",
"model.audio.layers.20.self_attn_layer_norm.bias",
"model.audio.layers.21.fc1",
"model.audio.layers.21.fc1.bias",
"model.audio.layers.21.fc2",
"model.audio.layers.21.fc2.bias",
"model.audio.layers.21.final_layer_norm",
"model.audio.layers.21.final_layer_norm.bias",
"model.audio.layers.21.self_attn.k_proj",
"model.audio.layers.21.self_attn.out_proj",
"model.audio.layers.21.self_attn.out_proj.bias",
"model.audio.layers.21.self_attn.q_proj",
"model.audio.layers.21.self_attn.q_proj.bias",
"model.audio.layers.21.self_attn.v_proj",
"model.audio.layers.21.self_attn.v_proj.bias",
"model.audio.layers.21.self_attn_layer_norm",
"model.audio.layers.21.self_attn_layer_norm.bias",
"model.audio.layers.22.fc1",
"model.audio.layers.22.fc1.bias",
"model.audio.layers.22.fc2",
"model.audio.layers.22.fc2.bias",
"model.audio.layers.22.final_layer_norm",
"model.audio.layers.22.final_layer_norm.bias",
"model.audio.layers.22.self_attn.k_proj",
"model.audio.layers.22.self_attn.out_proj",
"model.audio.layers.22.self_attn.out_proj.bias",
"model.audio.layers.22.self_attn.q_proj",
"model.audio.layers.22.self_attn.q_proj.bias",
"model.audio.layers.22.self_attn.v_proj",
"model.audio.layers.22.self_attn.v_proj.bias",
"model.audio.layers.22.self_attn_layer_norm",
"model.audio.layers.22.self_attn_layer_norm.bias",
"model.audio.layers.23.fc1",
"model.audio.layers.23.fc1.bias",
"model.audio.layers.23.fc2",
"model.audio.layers.23.fc2.bias",
"model.audio.layers.23.final_layer_norm",
"model.audio.layers.23.final_layer_norm.bias",
"model.audio.layers.23.self_attn.k_proj",
"model.audio.layers.23.self_attn.out_proj",
"model.audio.layers.23.self_attn.out_proj.bias",
"model.audio.layers.23.self_attn.q_proj",
"model.audio.layers.23.self_attn.q_proj.bias",
"model.audio.layers.23.self_attn.v_proj",
"model.audio.layers.23.self_attn.v_proj.bias",
"model.audio.layers.23.self_attn_layer_norm",
"model.audio.layers.23.self_attn_layer_norm.bias",
"model.audio.layers.24.fc1",
"model.audio.layers.24.fc1.bias",
"model.audio.layers.24.fc2",
"model.audio.layers.24.fc2.bias",
"model.audio.layers.24.final_layer_norm",
"model.audio.layers.24.final_layer_norm.bias",
"model.audio.layers.24.self_attn.k_proj",
"model.audio.layers.24.self_attn.out_proj",
"model.audio.layers.24.self_attn.out_proj.bias",
"model.audio.layers.24.self_attn.q_proj",
"model.audio.layers.24.self_attn.q_proj.bias",
"model.audio.layers.24.self_attn.v_proj",
"model.audio.layers.24.self_attn.v_proj.bias",
"model.audio.layers.24.self_attn_layer_norm",
"model.audio.layers.24.self_attn_layer_norm.bias",
"model.audio.layers.25.fc1",
"model.audio.layers.25.fc1.bias",
"model.audio.layers.25.fc2",
"model.audio.layers.25.fc2.bias",
"model.audio.layers.25.final_layer_norm",
"model.audio.layers.25.final_layer_norm.bias",
"model.audio.layers.25.self_attn.k_proj",
"model.audio.layers.25.self_attn.out_proj",
"model.audio.layers.25.self_attn.out_proj.bias",
"model.audio.layers.25.self_attn.q_proj",
"model.audio.layers.25.self_attn.q_proj.bias",
"model.audio.layers.25.self_attn.v_proj",
"model.audio.layers.25.self_attn.v_proj.bias",
"model.audio.layers.25.self_attn_layer_norm",
"model.audio.layers.25.self_attn_layer_norm.bias",
"model.audio.layers.26.fc1",
"model.audio.layers.26.fc1.bias",
"model.audio.layers.26.fc2",
"model.audio.layers.26.fc2.bias",
"model.audio.layers.26.final_layer_norm",
"model.audio.layers.26.final_layer_norm.bias",
"model.audio.layers.26.self_attn.k_proj",
"model.audio.layers.26.self_attn.out_proj",
"model.audio.layers.26.self_attn.out_proj.bias",
"model.audio.layers.26.self_attn.q_proj",
"model.audio.layers.26.self_attn.q_proj.bias",
"model.audio.layers.26.self_attn.v_proj",
"model.audio.layers.26.self_attn.v_proj.bias",
"model.audio.layers.26.self_attn_layer_norm",
"model.audio.layers.26.self_attn_layer_norm.bias",
"model.audio.layers.27.fc1",
"model.audio.layers.27.fc1.bias",
"model.audio.layers.27.fc2",
"model.audio.layers.27.fc2.bias",
"model.audio.layers.27.final_layer_norm",
"model.audio.layers.27.final_layer_norm.bias",
"model.audio.layers.27.self_attn.k_proj",
"model.audio.layers.27.self_attn.out_proj",
"model.audio.layers.27.self_attn.out_proj.bias",
"model.audio.layers.27.self_attn.q_proj",
"model.audio.layers.27.self_attn.q_proj.bias",
"model.audio.layers.27.self_attn.v_proj",
"model.audio.layers.27.self_attn.v_proj.bias",
"model.audio.layers.27.self_attn_layer_norm",
"model.audio.layers.27.self_attn_layer_norm.bias",
"model.audio.layers.28.fc1",
"model.audio.layers.28.fc1.bias",
"model.audio.layers.28.fc2",
"model.audio.layers.28.fc2.bias",
"model.audio.layers.28.final_layer_norm",
"model.audio.layers.28.final_layer_norm.bias",
"model.audio.layers.28.self_attn.k_proj",
"model.audio.layers.28.self_attn.out_proj",
"model.audio.layers.28.self_attn.out_proj.bias",
"model.audio.layers.28.self_attn.q_proj",
"model.audio.layers.28.self_attn.q_proj.bias",
"model.audio.layers.28.self_attn.v_proj",
"model.audio.layers.28.self_attn.v_proj.bias",
"model.audio.layers.28.self_attn_layer_norm",
"model.audio.layers.28.self_attn_layer_norm.bias",
"model.audio.layers.29.fc1",
"model.audio.layers.29.fc1.bias",
"model.audio.layers.29.fc2",
"model.audio.layers.29.fc2.bias",
"model.audio.layers.29.final_layer_norm",
"model.audio.layers.29.final_layer_norm.bias",
"model.audio.layers.29.self_attn.k_proj",
"model.audio.layers.29.self_attn.out_proj",
"model.audio.layers.29.self_attn.out_proj.bias",
"model.audio.layers.29.self_attn.q_proj",
"model.audio.layers.29.self_attn.q_proj.bias",
"model.audio.layers.29.self_attn.v_proj",
"model.audio.layers.29.self_attn.v_proj.bias",
"model.audio.layers.29.self_attn_layer_norm",
"model.audio.layers.29.self_attn_layer_norm.bias",
"model.audio.layers.3.fc1",
"model.audio.layers.3.fc1.bias",
"model.audio.layers.3.fc2",
"model.audio.layers.3.fc2.bias",
"model.audio.layers.3.final_layer_norm",
"model.audio.layers.3.final_layer_norm.bias",
"model.audio.layers.3.self_attn.k_proj",
"model.audio.layers.3.self_attn.out_proj",
"model.audio.layers.3.self_attn.out_proj.bias",
"model.audio.layers.3.self_attn.q_proj",
"model.audio.layers.3.self_attn.q_proj.bias",
"model.audio.layers.3.self_attn.v_proj",
"model.audio.layers.3.self_attn.v_proj.bias",
"model.audio.layers.3.self_attn_layer_norm",
"model.audio.layers.3.self_attn_layer_norm.bias",
"model.audio.layers.30.fc1",
"model.audio.layers.30.fc1.bias",
"model.audio.layers.30.fc2",
"model.audio.layers.30.fc2.bias",
"model.audio.layers.30.final_layer_norm",
"model.audio.layers.30.final_layer_norm.bias",
"model.audio.layers.30.self_attn.k_proj",
"model.audio.layers.30.self_attn.out_proj",
"model.audio.layers.30.self_attn.out_proj.bias",
"model.audio.layers.30.self_attn.q_proj",
"model.audio.layers.30.self_attn.q_proj.bias",
"model.audio.layers.30.self_attn.v_proj",
"model.audio.layers.30.self_attn.v_proj.bias",
"model.audio.layers.30.self_attn_layer_norm",
"model.audio.layers.30.self_attn_layer_norm.bias",
"model.audio.layers.31.fc1",
"model.audio.layers.31.fc1.bias",
"model.audio.layers.31.fc2",
"model.audio.layers.31.fc2.bias",
"model.audio.layers.31.final_layer_norm",
"model.audio.layers.31.final_layer_norm.bias",
"model.audio.layers.31.self_attn.k_proj",
"model.audio.layers.31.self_attn.out_proj",
"model.audio.layers.31.self_attn.out_proj.bias",
"model.audio.layers.31.self_attn.q_proj",
"model.audio.layers.31.self_attn.q_proj.bias",
"model.audio.layers.31.self_attn.v_proj",
"model.audio.layers.31.self_attn.v_proj.bias",
"model.audio.layers.31.self_attn_layer_norm",
"model.audio.layers.31.self_attn_layer_norm.bias",
"model.audio.layers.4.fc1",
"model.audio.layers.4.fc1.bias",
"model.audio.layers.4.fc2",
"model.audio.layers.4.fc2.bias",
"model.audio.layers.4.final_layer_norm",
"model.audio.layers.4.final_layer_norm.bias",
"model.audio.layers.4.self_attn.k_proj",
"model.audio.layers.4.self_attn.out_proj",
"model.audio.layers.4.self_attn.out_proj.bias",
"model.audio.layers.4.self_attn.q_proj",
"model.audio.layers.4.self_attn.q_proj.bias",
"model.audio.layers.4.self_attn.v_proj",
"model.audio.layers.4.self_attn.v_proj.bias",
"model.audio.layers.4.self_attn_layer_norm",
"model.audio.layers.4.self_attn_layer_norm.bias",
"model.audio.layers.5.fc1",
"model.audio.layers.5.fc1.bias",
"model.audio.layers.5.fc2",
"model.audio.layers.5.fc2.bias",
"model.audio.layers.5.final_layer_norm",
"model.audio.layers.5.final_layer_norm.bias",
"model.audio.layers.5.self_attn.k_proj",
"model.audio.layers.5.self_attn.out_proj",
"model.audio.layers.5.self_attn.out_proj.bias",
"model.audio.layers.5.self_attn.q_proj",
"model.audio.layers.5.self_attn.q_proj.bias",
"model.audio.layers.5.self_attn.v_proj",
"model.audio.layers.5.self_attn.v_proj.bias",
"model.audio.layers.5.self_attn_layer_norm",
"model.audio.layers.5.self_attn_layer_norm.bias",
"model.audio.layers.6.fc1",
"model.audio.layers.6.fc1.bias",
"model.audio.layers.6.fc2",
"model.audio.layers.6.fc2.bias",
"model.audio.layers.6.final_layer_norm",
"model.audio.layers.6.final_layer_norm.bias",
"model.audio.layers.6.self_attn.k_proj",
"model.audio.layers.6.self_attn.out_proj",
"model.audio.layers.6.self_attn.out_proj.bias",
"model.audio.layers.6.self_attn.q_proj",
"model.audio.layers.6.self_attn.q_proj.bias",
"model.audio.layers.6.self_attn.v_proj",
"model.audio.layers.6.self_attn.v_proj.bias",
"model.audio.layers.6.self_attn_layer_norm",
"model.audio.layers.6.self_attn_layer_norm.bias",
"model.audio.layers.7.fc1",
"model.audio.layers.7.fc1.bias",
"model.audio.layers.7.fc2",
"model.audio.layers.7.fc2.bias",
"model.audio.layers.7.final_layer_norm",
"model.audio.layers.7.final_layer_norm.bias",
"model.audio.layers.7.self_attn.k_proj",
"model.audio.layers.7.self_attn.out_proj",
"model.audio.layers.7.self_attn.out_proj.bias",
"model.audio.layers.7.self_attn.q_proj",
"model.audio.layers.7.self_attn.q_proj.bias",
"model.audio.layers.7.self_attn.v_proj",
"model.audio.layers.7.self_attn.v_proj.bias",
"model.audio.layers.7.self_attn_layer_norm",
"model.audio.layers.7.self_attn_layer_norm.bias",
"model.audio.layers.8.fc1",
"model.audio.layers.8.fc1.bias",
"model.audio.layers.8.fc2",
"model.audio.layers.8.fc2.bias",
"model.audio.layers.8.final_layer_norm",
"model.audio.layers.8.final_layer_norm.bias",
"model.audio.layers.8.self_attn.k_proj",
"model.audio.layers.8.self_attn.out_proj",
"model.audio.layers.8.self_attn.out_proj.bias",
"model.audio.layers.8.self_attn.q_proj",
"model.audio.layers.8.self_attn.q_proj.bias",
"model.audio.layers.8.self_attn.v_proj",
"model.audio.layers.8.self_attn.v_proj.bias",
"model.audio.layers.8.self_attn_layer_norm",
"model.audio.layers.8.self_attn_layer_norm.bias",
"model.audio.layers.9.fc1",
"model.audio.layers.9.fc1.bias",
"model.audio.layers.9.fc2",
"model.audio.layers.9.fc2.bias",
"model.audio.layers.9.final_layer_norm",
"model.audio.layers.9.final_layer_norm.bias",
"model.audio.layers.9.self_attn.k_proj",
"model.audio.layers.9.self_attn.out_proj",
"model.audio.layers.9.self_attn.out_proj.bias",
"model.audio.layers.9.self_attn.q_proj",
"model.audio.layers.9.self_attn.q_proj.bias",
"model.audio.layers.9.self_attn.v_proj",
"model.audio.layers.9.self_attn.v_proj.bias",
"model.audio.layers.9.self_attn_layer_norm",
"model.audio.layers.9.self_attn_layer_norm.bias",
"model.audio.q_tokens",
"model.audio.qformer.bert.embeddings.LayerNorm.beta",
"model.audio.qformer.bert.embeddings.LayerNorm.gamma",
"model.audio.qformer.bert.embeddings.position_ids",
"model.audio.qformer.bert.encoder.layer.0.attention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.0.attention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.0.attention.output.dense",
"model.audio.qformer.bert.encoder.layer.0.attention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.0.attention.self.key",
"model.audio.qformer.bert.encoder.layer.0.attention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.0.attention.self.query",
"model.audio.qformer.bert.encoder.layer.0.attention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.0.attention.self.value",
"model.audio.qformer.bert.encoder.layer.0.attention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.dense",
"model.audio.qformer.bert.encoder.layer.0.crossattention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.key",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.query",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.value",
"model.audio.qformer.bert.encoder.layer.0.crossattention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.0.intermediate_query.dense",
"model.audio.qformer.bert.encoder.layer.0.intermediate_query.dense.bias",
"model.audio.qformer.bert.encoder.layer.0.output_query.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.0.output_query.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.0.output_query.dense",
"model.audio.qformer.bert.encoder.layer.0.output_query.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.1.attention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.1.attention.output.dense",
"model.audio.qformer.bert.encoder.layer.1.attention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.self.key",
"model.audio.qformer.bert.encoder.layer.1.attention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.self.query",
"model.audio.qformer.bert.encoder.layer.1.attention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.1.attention.self.value",
"model.audio.qformer.bert.encoder.layer.1.attention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.dense",
"model.audio.qformer.bert.encoder.layer.1.crossattention.output.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.key",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.key.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.query",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.query.bias",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.value",
"model.audio.qformer.bert.encoder.layer.1.crossattention.self.value.bias",
"model.audio.qformer.bert.encoder.layer.1.intermediate_query.dense",
"model.audio.qformer.bert.encoder.layer.1.intermediate_query.dense.bias",
"model.audio.qformer.bert.encoder.layer.1.output_query.LayerNorm.beta",
"model.audio.qformer.bert.encoder.layer.1.output_query.LayerNorm.gamma",
"model.audio.qformer.bert.encoder.layer.1.output_query.dense",
"model.audio.qformer.bert.encoder.layer.1.output_query.dense.bias",
"model.language_model.embed_tokens",
"model.language_model.layers.0.linear_attn.conv1d",
"model.language_model.layers.0.linear_attn.in_proj_a",
"model.language_model.layers.0.linear_attn.in_proj_b",
"model.language_model.layers.1.linear_attn.conv1d",
"model.language_model.layers.1.linear_attn.in_proj_a",
"model.language_model.layers.1.linear_attn.in_proj_b",
"model.language_model.layers.10.linear_attn.conv1d",
"model.language_model.layers.10.linear_attn.in_proj_a",
"model.language_model.layers.10.linear_attn.in_proj_b",
"model.language_model.layers.12.linear_attn.conv1d",
"model.language_model.layers.12.linear_attn.in_proj_a",
"model.language_model.layers.12.linear_attn.in_proj_b",
"model.language_model.layers.13.linear_attn.conv1d",
"model.language_model.layers.13.linear_attn.in_proj_a",
"model.language_model.layers.13.linear_attn.in_proj_b",
"model.language_model.layers.14.linear_attn.conv1d",
"model.language_model.layers.14.linear_attn.in_proj_a",
"model.language_model.layers.14.linear_attn.in_proj_b",
"model.language_model.layers.16.linear_attn.conv1d",
"model.language_model.layers.16.linear_attn.in_proj_a",
"model.language_model.layers.16.linear_attn.in_proj_b",
"model.language_model.layers.17.linear_attn.conv1d",
"model.language_model.layers.17.linear_attn.in_proj_a",
"model.language_model.layers.17.linear_attn.in_proj_b",
"model.language_model.layers.18.linear_attn.conv1d",
"model.language_model.layers.18.linear_attn.in_proj_a",
"model.language_model.layers.18.linear_attn.in_proj_b",
"model.language_model.layers.2.linear_attn.conv1d",
"model.language_model.layers.2.linear_attn.in_proj_a",
"model.language_model.layers.2.linear_attn.in_proj_b",
"model.language_model.layers.20.linear_attn.conv1d",
"model.language_model.layers.20.linear_attn.in_proj_a",
"model.language_model.layers.20.linear_attn.in_proj_b",
"model.language_model.layers.21.linear_attn.conv1d",
"model.language_model.layers.21.linear_attn.in_proj_a",
"model.language_model.layers.21.linear_attn.in_proj_b",
"model.language_model.layers.22.linear_attn.conv1d",
"model.language_model.layers.22.linear_attn.in_proj_a",
"model.language_model.layers.22.linear_attn.in_proj_b",
"model.language_model.layers.24.linear_attn.conv1d",
"model.language_model.layers.24.linear_attn.in_proj_a",
"model.language_model.layers.24.linear_attn.in_proj_b",
"model.language_model.layers.25.linear_attn.conv1d",
"model.language_model.layers.25.linear_attn.in_proj_a",
"model.language_model.layers.25.linear_attn.in_proj_b",
"model.language_model.layers.26.linear_attn.conv1d",
"model.language_model.layers.26.linear_attn.in_proj_a",
"model.language_model.layers.26.linear_attn.in_proj_b",
"model.language_model.layers.28.linear_attn.conv1d",
"model.language_model.layers.28.linear_attn.in_proj_a",
"model.language_model.layers.28.linear_attn.in_proj_b",
"model.language_model.layers.29.linear_attn.conv1d",
"model.language_model.layers.29.linear_attn.in_proj_a",
"model.language_model.layers.29.linear_attn.in_proj_b",
"model.language_model.layers.30.linear_attn.conv1d",
"model.language_model.layers.30.linear_attn.in_proj_a",
"model.language_model.layers.30.linear_attn.in_proj_b",
"model.language_model.layers.4.linear_attn.conv1d",
"model.language_model.layers.4.linear_attn.in_proj_a",
"model.language_model.layers.4.linear_attn.in_proj_b",
"model.language_model.layers.5.linear_attn.conv1d",
"model.language_model.layers.5.linear_attn.in_proj_a",
"model.language_model.layers.5.linear_attn.in_proj_b",
"model.language_model.layers.6.linear_attn.conv1d",
"model.language_model.layers.6.linear_attn.in_proj_a",
"model.language_model.layers.6.linear_attn.in_proj_b",
"model.language_model.layers.8.linear_attn.conv1d",
"model.language_model.layers.8.linear_attn.in_proj_a",
"model.language_model.layers.8.linear_attn.in_proj_b",
"model.language_model.layers.9.linear_attn.conv1d",
"model.language_model.layers.9.linear_attn.in_proj_a",
"model.language_model.layers.9.linear_attn.in_proj_b",
"model.language_model.norm",
"model.visual",
"model.visual.blocks.0.attn.proj",
"model.visual.blocks.0.attn.proj.bias",
"model.visual.blocks.0.attn.qkv",
"model.visual.blocks.0.attn.qkv.bias",
"model.visual.blocks.0.mlp.linear_fc1",
"model.visual.blocks.0.mlp.linear_fc1.bias",
"model.visual.blocks.0.mlp.linear_fc2",
"model.visual.blocks.0.mlp.linear_fc2.bias",
"model.visual.blocks.0.norm1",
"model.visual.blocks.0.norm1.bias",
"model.visual.blocks.0.norm2",
"model.visual.blocks.0.norm2.bias",
"model.visual.blocks.1.attn.proj",
"model.visual.blocks.1.attn.proj.bias",
"model.visual.blocks.1.attn.qkv",
"model.visual.blocks.1.attn.qkv.bias",
"model.visual.blocks.1.mlp.linear_fc1",
"model.visual.blocks.1.mlp.linear_fc1.bias",
"model.visual.blocks.1.mlp.linear_fc2",
"model.visual.blocks.1.mlp.linear_fc2.bias",
"model.visual.blocks.1.norm1",
"model.visual.blocks.1.norm1.bias",
"model.visual.blocks.1.norm2",
"model.visual.blocks.1.norm2.bias",
"model.visual.blocks.10.attn.proj",
"model.visual.blocks.10.attn.proj.bias",
"model.visual.blocks.10.attn.qkv",
"model.visual.blocks.10.attn.qkv.bias",
"model.visual.blocks.10.mlp.linear_fc1",
"model.visual.blocks.10.mlp.linear_fc1.bias",
"model.visual.blocks.10.mlp.linear_fc2",
"model.visual.blocks.10.mlp.linear_fc2.bias",
"model.visual.blocks.10.norm1",
"model.visual.blocks.10.norm1.bias",
"model.visual.blocks.10.norm2",
"model.visual.blocks.10.norm2.bias",
"model.visual.blocks.11.attn.proj",
"model.visual.blocks.11.attn.proj.bias",
"model.visual.blocks.11.attn.qkv",
"model.visual.blocks.11.attn.qkv.bias",
"model.visual.blocks.11.mlp.linear_fc1",
"model.visual.blocks.11.mlp.linear_fc1.bias",
"model.visual.blocks.11.mlp.linear_fc2",
"model.visual.blocks.11.mlp.linear_fc2.bias",
"model.visual.blocks.11.norm1",
"model.visual.blocks.11.norm1.bias",
"model.visual.blocks.11.norm2",
"model.visual.blocks.11.norm2.bias",
"model.visual.blocks.12.attn.proj",
"model.visual.blocks.12.attn.proj.bias",
"model.visual.blocks.12.attn.qkv",
"model.visual.blocks.12.attn.qkv.bias",
"model.visual.blocks.12.mlp.linear_fc1",
"model.visual.blocks.12.mlp.linear_fc1.bias",
"model.visual.blocks.12.mlp.linear_fc2",
"model.visual.blocks.12.mlp.linear_fc2.bias",
"model.visual.blocks.12.norm1",
"model.visual.blocks.12.norm1.bias",
"model.visual.blocks.12.norm2",
"model.visual.blocks.12.norm2.bias",
"model.visual.blocks.13.attn.proj",
"model.visual.blocks.13.attn.proj.bias",
"model.visual.blocks.13.attn.qkv",
"model.visual.blocks.13.attn.qkv.bias",
"model.visual.blocks.13.mlp.linear_fc1",
"model.visual.blocks.13.mlp.linear_fc1.bias",
"model.visual.blocks.13.mlp.linear_fc2",
"model.visual.blocks.13.mlp.linear_fc2.bias",
"model.visual.blocks.13.norm1",
"model.visual.blocks.13.norm1.bias",
"model.visual.blocks.13.norm2",
"model.visual.blocks.13.norm2.bias",
"model.visual.blocks.14.attn.proj",
"model.visual.blocks.14.attn.proj.bias",
"model.visual.blocks.14.attn.qkv",
"model.visual.blocks.14.attn.qkv.bias",
"model.visual.blocks.14.mlp.linear_fc1",
"model.visual.blocks.14.mlp.linear_fc1.bias",
"model.visual.blocks.14.mlp.linear_fc2",
"model.visual.blocks.14.mlp.linear_fc2.bias",
"model.visual.blocks.14.norm1",
"model.visual.blocks.14.norm1.bias",
"model.visual.blocks.14.norm2",
"model.visual.blocks.14.norm2.bias",
"model.visual.blocks.15.attn.proj",
"model.visual.blocks.15.attn.proj.bias",
"model.visual.blocks.15.attn.qkv",
"model.visual.blocks.15.attn.qkv.bias",
"model.visual.blocks.15.mlp.linear_fc1",
"model.visual.blocks.15.mlp.linear_fc1.bias",
"model.visual.blocks.15.mlp.linear_fc2",
"model.visual.blocks.15.mlp.linear_fc2.bias",
"model.visual.blocks.15.norm1",
"model.visual.blocks.15.norm1.bias",
"model.visual.blocks.15.norm2",
"model.visual.blocks.15.norm2.bias",
"model.visual.blocks.16.attn.proj",
"model.visual.blocks.16.attn.proj.bias",
"model.visual.blocks.16.attn.qkv",
"model.visual.blocks.16.attn.qkv.bias",
"model.visual.blocks.16.mlp.linear_fc1",
"model.visual.blocks.16.mlp.linear_fc1.bias",
"model.visual.blocks.16.mlp.linear_fc2",
"model.visual.blocks.16.mlp.linear_fc2.bias",
"model.visual.blocks.16.norm1",
"model.visual.blocks.16.norm1.bias",
"model.visual.blocks.16.norm2",
"model.visual.blocks.16.norm2.bias",
"model.visual.blocks.17.attn.proj",
"model.visual.blocks.17.attn.proj.bias",
"model.visual.blocks.17.attn.qkv",
"model.visual.blocks.17.attn.qkv.bias",
"model.visual.blocks.17.mlp.linear_fc1",
"model.visual.blocks.17.mlp.linear_fc1.bias",
"model.visual.blocks.17.mlp.linear_fc2",
"model.visual.blocks.17.mlp.linear_fc2.bias",
"model.visual.blocks.17.norm1",
"model.visual.blocks.17.norm1.bias",
"model.visual.blocks.17.norm2",
"model.visual.blocks.17.norm2.bias",
"model.visual.blocks.18.attn.proj",
"model.visual.blocks.18.attn.proj.bias",
"model.visual.blocks.18.attn.qkv",
"model.visual.blocks.18.attn.qkv.bias",
"model.visual.blocks.18.mlp.linear_fc1",
"model.visual.blocks.18.mlp.linear_fc1.bias",
"model.visual.blocks.18.mlp.linear_fc2",
"model.visual.blocks.18.mlp.linear_fc2.bias",
"model.visual.blocks.18.norm1",
"model.visual.blocks.18.norm1.bias",
"model.visual.blocks.18.norm2",
"model.visual.blocks.18.norm2.bias",
"model.visual.blocks.19.attn.proj",
"model.visual.blocks.19.attn.proj.bias",
"model.visual.blocks.19.attn.qkv",
"model.visual.blocks.19.attn.qkv.bias",
"model.visual.blocks.19.mlp.linear_fc1",
"model.visual.blocks.19.mlp.linear_fc1.bias",
"model.visual.blocks.19.mlp.linear_fc2",
"model.visual.blocks.19.mlp.linear_fc2.bias",
"model.visual.blocks.19.norm1",
"model.visual.blocks.19.norm1.bias",
"model.visual.blocks.19.norm2",
"model.visual.blocks.19.norm2.bias",
"model.visual.blocks.2.attn.proj",
"model.visual.blocks.2.attn.proj.bias",
"model.visual.blocks.2.attn.qkv",
"model.visual.blocks.2.attn.qkv.bias",
"model.visual.blocks.2.mlp.linear_fc1",
"model.visual.blocks.2.mlp.linear_fc1.bias",
"model.visual.blocks.2.mlp.linear_fc2",
"model.visual.blocks.2.mlp.linear_fc2.bias",
"model.visual.blocks.2.norm1",
"model.visual.blocks.2.norm1.bias",
"model.visual.blocks.2.norm2",
"model.visual.blocks.2.norm2.bias",
"model.visual.blocks.20.attn.proj",
"model.visual.blocks.20.attn.proj.bias",
"model.visual.blocks.20.attn.qkv",
"model.visual.blocks.20.attn.qkv.bias",
"model.visual.blocks.20.mlp.linear_fc1",
"model.visual.blocks.20.mlp.linear_fc1.bias",
"model.visual.blocks.20.mlp.linear_fc2",
"model.visual.blocks.20.mlp.linear_fc2.bias",
"model.visual.blocks.20.norm1",
"model.visual.blocks.20.norm1.bias",
"model.visual.blocks.20.norm2",
"model.visual.blocks.20.norm2.bias",
"model.visual.blocks.21.attn.proj",
"model.visual.blocks.21.attn.proj.bias",
"model.visual.blocks.21.attn.qkv",
"model.visual.blocks.21.attn.qkv.bias",
"model.visual.blocks.21.mlp.linear_fc1",
"model.visual.blocks.21.mlp.linear_fc1.bias",
"model.visual.blocks.21.mlp.linear_fc2",
"model.visual.blocks.21.mlp.linear_fc2.bias",
"model.visual.blocks.21.norm1",
"model.visual.blocks.21.norm1.bias",
"model.visual.blocks.21.norm2",
"model.visual.blocks.21.norm2.bias",
"model.visual.blocks.22.attn.proj",
"model.visual.blocks.22.attn.proj.bias",
"model.visual.blocks.22.attn.qkv",
"model.visual.blocks.22.attn.qkv.bias",
"model.visual.blocks.22.mlp.linear_fc1",
"model.visual.blocks.22.mlp.linear_fc1.bias",
"model.visual.blocks.22.mlp.linear_fc2",
"model.visual.blocks.22.mlp.linear_fc2.bias",
"model.visual.blocks.22.norm1",
"model.visual.blocks.22.norm1.bias",
"model.visual.blocks.22.norm2",
"model.visual.blocks.22.norm2.bias",
"model.visual.blocks.23.attn.proj",
"model.visual.blocks.23.attn.proj.bias",
"model.visual.blocks.23.attn.qkv",
"model.visual.blocks.23.attn.qkv.bias",
"model.visual.blocks.23.mlp.linear_fc1",
"model.visual.blocks.23.mlp.linear_fc1.bias",
"model.visual.blocks.23.mlp.linear_fc2",
"model.visual.blocks.23.mlp.linear_fc2.bias",
"model.visual.blocks.23.norm1",
"model.visual.blocks.23.norm1.bias",
"model.visual.blocks.23.norm2",
"model.visual.blocks.23.norm2.bias",
"model.visual.blocks.3.attn.proj",
"model.visual.blocks.3.attn.proj.bias",
"model.visual.blocks.3.attn.qkv",
"model.visual.blocks.3.attn.qkv.bias",
"model.visual.blocks.3.mlp.linear_fc1",
"model.visual.blocks.3.mlp.linear_fc1.bias",
"model.visual.blocks.3.mlp.linear_fc2",
"model.visual.blocks.3.mlp.linear_fc2.bias",
"model.visual.blocks.3.norm1",
"model.visual.blocks.3.norm1.bias",
"model.visual.blocks.3.norm2",
"model.visual.blocks.3.norm2.bias",
"model.visual.blocks.4.attn.proj",
"model.visual.blocks.4.attn.proj.bias",
"model.visual.blocks.4.attn.qkv",
"model.visual.blocks.4.attn.qkv.bias",
"model.visual.blocks.4.mlp.linear_fc1",
"model.visual.blocks.4.mlp.linear_fc1.bias",
"model.visual.blocks.4.mlp.linear_fc2",
"model.visual.blocks.4.mlp.linear_fc2.bias",
"model.visual.blocks.4.norm1",
"model.visual.blocks.4.norm1.bias",
"model.visual.blocks.4.norm2",
"model.visual.blocks.4.norm2.bias",
"model.visual.blocks.5.attn.proj",
"model.visual.blocks.5.attn.proj.bias",
"model.visual.blocks.5.attn.qkv",
"model.visual.blocks.5.attn.qkv.bias",
"model.visual.blocks.5.mlp.linear_fc1",
"model.visual.blocks.5.mlp.linear_fc1.bias",
"model.visual.blocks.5.mlp.linear_fc2",
"model.visual.blocks.5.mlp.linear_fc2.bias",
"model.visual.blocks.5.norm1",
"model.visual.blocks.5.norm1.bias",
"model.visual.blocks.5.norm2",
"model.visual.blocks.5.norm2.bias",
"model.visual.blocks.6.attn.proj",
"model.visual.blocks.6.attn.proj.bias",
"model.visual.blocks.6.attn.qkv",
"model.visual.blocks.6.attn.qkv.bias",
"model.visual.blocks.6.mlp.linear_fc1",
"model.visual.blocks.6.mlp.linear_fc1.bias",
"model.visual.blocks.6.mlp.linear_fc2",
"model.visual.blocks.6.mlp.linear_fc2.bias",
"model.visual.blocks.6.norm1",
"model.visual.blocks.6.norm1.bias",
"model.visual.blocks.6.norm2",
"model.visual.blocks.6.norm2.bias",
"model.visual.blocks.7.attn.proj",
"model.visual.blocks.7.attn.proj.bias",
"model.visual.blocks.7.attn.qkv",
"model.visual.blocks.7.attn.qkv.bias",
"model.visual.blocks.7.mlp.linear_fc1",
"model.visual.blocks.7.mlp.linear_fc1.bias",
"model.visual.blocks.7.mlp.linear_fc2",
"model.visual.blocks.7.mlp.linear_fc2.bias",
"model.visual.blocks.7.norm1",
"model.visual.blocks.7.norm1.bias",
"model.visual.blocks.7.norm2",
"model.visual.blocks.7.norm2.bias",
"model.visual.blocks.8.attn.proj",
"model.visual.blocks.8.attn.proj.bias",
"model.visual.blocks.8.attn.qkv",
"model.visual.blocks.8.attn.qkv.bias",
"model.visual.blocks.8.mlp.linear_fc1",
"model.visual.blocks.8.mlp.linear_fc1.bias",
"model.visual.blocks.8.mlp.linear_fc2",
"model.visual.blocks.8.mlp.linear_fc2.bias",
"model.visual.blocks.8.norm1",
"model.visual.blocks.8.norm1.bias",
"model.visual.blocks.8.norm2",
"model.visual.blocks.8.norm2.bias",
"model.visual.blocks.9.attn.proj",
"model.visual.blocks.9.attn.proj.bias",
"model.visual.blocks.9.attn.qkv",
"model.visual.blocks.9.attn.qkv.bias",
"model.visual.blocks.9.mlp.linear_fc1",
"model.visual.blocks.9.mlp.linear_fc1.bias",
"model.visual.blocks.9.mlp.linear_fc2",
"model.visual.blocks.9.mlp.linear_fc2.bias",
"model.visual.blocks.9.norm1",
"model.visual.blocks.9.norm1.bias",
"model.visual.blocks.9.norm2",
"model.visual.blocks.9.norm2.bias",
"model.visual.merger.linear_fc1",
"model.visual.merger.linear_fc1.bias",
"model.visual.merger.linear_fc2",
"model.visual.merger.linear_fc2.bias",
"model.visual.merger.norm",
"model.visual.merger.norm.bias",
"model.visual.patch_embed.proj",
"model.visual.patch_embed.proj.bias",
"model.visual.pos_embed",
"mtp"
]
},
"quantized_linear_matrices": 200,
"architecture": {
"family": "dense",
"model_type": "qwen3_5",
"layers": 32,
"experts": 0,
"hidden_size": 2560,
"intermediate_size": 9216,
"shared_expert_intermediate_size": null,
"layer_types": [
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention"
],
"linear_attention_layers": 24,
"full_attention_layers": 8,
"model_class": "modeling_wemm.WeMMVideoForConditionalGeneration"
},
"quantized_expert_matrices": 0,
"protected_gdn_in_proj_a_b": 48,
"protected_tensor_count": 1068,
"protected_tensors_bit_exact": true,
"independent_final_file_audit": {
"completed": true,
"tensor_count": 1468,
"sha256": "bd5ed7eaf0682ee781175a249c9f86b9366d0d31980f72409ec6b7d939486ef4",
"bytes": 7659403688,
"all_tensor_hashes_reloaded": true
},
"native_inference_validated": false,
"native_low_precision_gemm_validated": false,
"published": false,
"versions": {
"torch": "2.6.0+cu124",
"transformers": "5.5.3"
},
"conversion_elapsed_seconds": 97.74387492053211
}
{
"status": "quantized_and_independently_reloaded_native_inference_pending",
"algorithm": "Transformers 5.5.3 Fp8Quantize.convert",
"converter_sha256": "9baa1fdc55581355dabaf10d68ddafcd53983fe8736737c991c2ea78b03a3d9f",
"source": "tencent/WeMM-Video-4B",
"source_files": [
{
"path": "tencent/WeMM-Video-4B/model-00001-of-00003.safetensors",
"bytes": 4306667392,
"sha256": "43adb5f51324177c3c40661982f5729b2c927b8c2f3eba9352ece3e0d5b2e2fd"
},
{
"path": "tencent/WeMM-Video-4B/model-00002-of-00003.safetensors",
"bytes": 4331122608,
"sha256": "c271269cd9ad157ace651e2f6ee2f28c8ee80c52324f8979ddb138ba653f069c"
},
{
"path": "tencent/WeMM-Video-4B/model-00003-of-00003.safetensors",
"bytes": 2585872480,
"sha256": "b9ecafb62c8c758242c926830ed628ef685c42e0be6b779c05d1eff2da1fdfd9"
}
],
"source_config_sha256": "5eca5c5955cdc4ec20ddb433053130c690304a0a26c2dcd031daef032b7235a4",
"official_repo": "Qwen/Qwen3.5-35B-A3B-FP8",
"official_revision": "9d1823d2dee688a6b25e77009dc727688c44936e",
"recipe_sha256": "1fa272f54cc586827ffa0e6d0ce14805d298adab4c9bd78663967bef155efda7",
"files": {
"wemm-video-capabilities-compact.png": {
"bytes": 2718998,
"sha256": "57d892abb395c639078b49ecf317698f038303adf1fc976ddf13ff9e9273c77f"
},
"wemm-video-architecture.png": {
"bytes": 116917,
"sha256": "9902c081a70abeaacf34c898c86c44587eebe0fd62f1923225d1726766eb0a7a"
},
"chat_template.jinja": {
"bytes": 7755,
"sha256": "273d8e0e683b885071fb17e08d71e5f2a5ddfb5309756181681de4f5a1822d80"
},
"config.json": {
"bytes": 50625,
"sha256": "77201ba25b2b6655411d8a5df5c03b51d730d140f89f82c826e11e5105ee027e"
},
"configuration_wemm.py": {
"bytes": 37064,
"sha256": "43ee3ec8477c6677f06835a893a146d18024aea0145086ceca3b6039e2d91887"
},
"generation_config.json": {
"bytes": 138,
"sha256": "b91f22ecb30192b6919660eb61c01d7602f93ff6764556b1991c71fbe9ca82e6"
},
"h5_modeling_bert.py": {
"bytes": 50024,
"sha256": "b2eac797422294be10295c27f5709c8fda362c55d6f476dcebfd5f38aa65c6dd"
},
"h5_modeling_whisper.py": {
"bytes": 69582,
"sha256": "b507884daee13857c17d2d97290183e71a5a1044253a49c5a9cb109ba3332907"
},
"model.safetensors": {
"bytes": 7659403688,
"sha256": "bd5ed7eaf0682ee781175a249c9f86b9366d0d31980f72409ec6b7d939486ef4"
},
"modeling_wemm.py": {
"bytes": 103798,
"sha256": "30c9e34d16f4b3720410c47034b2b807a3686c88f4558d5aab6c102fc1db0bde"
},
"official_qwen35_fp8_config.json": {
"bytes": 19172,
"sha256": "cc35eadc02b2a69542c59233f5d80cd89029234711ac11811afc46d57701f7b9"
},
"preprocessor_config.json": {
"bytes": 390,
"sha256": "27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516"
},
"processing_wemm.py": {
"bytes": 24880,
"sha256": "8f907439e011c47293a851e28efec1501c36b862ba8c04784d23a0a4a5aa36a8"
},
"processor_config.json": {
"bytes": 2734,
"sha256": "7067f0990504250c037bc718650754575b9002133b247d510ca86bf475b30d29"
},
"tokenizer.json": {
"bytes": 19989343,
"sha256": "510fd6ff51823f6b1c820595b5eed750b8ff0ee85961d2f330623b3cf5ff4019"
},
"tokenizer_config.json": {
"bytes": 1098,
"sha256": "83febe739ecefd99f984f546dfdc10149cedf293b9ee0291b99894341188b613"
},
"video_preprocessor_config.json": {
"bytes": 385,
"sha256": "7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13"
},
"wemm_media.py": {
"bytes": 11378,
"sha256": "2e4b427e3ddabc526f7984066a982bfa7c0a4d56f3d7c979fc41651463461122"
},
"wemm_video_policy.py": {